Major depression (MDD) is a highly prevalent disorder associated with significant morbidity and mortality and is estimated to be one of the leading causes of disability worldwide. The DSM-IV defined syndrome of MDD is a heterogeneous disorder characterized by a range of distinctive genetic, neural, and neuroendocrine diatheses and abnormalities. At the same time, individuals live and grow within an environment that influences developmental neurobiology, neurochemistry, and patterns of thought, feeling and behavior that further impact their vulnerability to the development of MDD. A variety of antidepressant drugs, psychotherapies, and non- drug somatic therapies have been demonstrated to be efficacious in acute treatment trials. The majority of patients in these trials, however, do not attain remission, increasing the risk for the development of chronic depression, suicide, substance abuse and several serious medical disorders. A major unmet need is the identification of predictors of remission to treatments, as has been utilized in other branches of medicine (e.g., oncology and infectious disease) to improve patient outcome. In view of advances in functional brain imaging, molecular and cellular neurobiology, genetics, and personality disorders and a number of promising findings in small studies, it is propitious to conduct studies to determine whether a concatenation of these factors predict antidepressant treatment remission as well as relapse and recurrence. In this application, we propose two studies. The first study is an expansion of our recently funded NIMH CIDAR 3-arm, 12-week treatment trial; we will increase the CIDAR sample size from 400 to 600 and follow all remitted patients for a total of two years. In this expanded study, 600 treatment nave adult depressed patients will be randomized to one of the following treatments: 1) escitalopram, an SSRI; 2) duloxetine, an SNRI; and 3) CBT. The primary goal of the CIDAR is to employ a multivariate approach to predict remission following acute treatment with a single therapy. The primary purpose of study of this proposal is to identify predictors of relapse and recurrence during the 21 months following remission to these three monotherapy treatments. The second study is designed to study prediction of short-term and long-term effects of additional pharmacotherapy and psychotherapy combination treatments for those who fail to remit with monotherapy. Our group has made a number of advances in identifying imaging procedures (e.g., resting BOLD fMRI), genetic polymorphisms, and personality disorder measures that predict remission, relapse, and recurrence of MDD. In addition, we will employ additional state- of-the-science measures of MDD and its treatment. Statistically, we will seek to determine which measure and/or combination of measures leads to prediction of remission, relapse, and recurrence to the treatments under study. Delineation of predictors of treatment remission, relapse, and recurrence of MDD will dramatically improve patient outcomes and reduce the risk of inadequate treatment. Major Depressive Disorder has a lifetime occurrence rate of 16% and is among the most frequent and debilitating of all medical disorders. Despite considerable advances in understanding the causes, nature, and treatment of depression, we still do not know which treatment will work for an individual patient. As a result of the proposed studies, it is hoped that the community clinician of tomorrow will be able to prescribe a specific treatment for an individual patient with major depression, with confidence that the prescribed treatment will provide effective and lasting relief from the symptoms of depression.